# The Mental Health Score That Did Not Move

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-20 | Updated: 2026-10-30

Canonical page: https://glpwatchdog.org/research/the-mental-health-score-that-did-not-move

> In the head-to-head trial, physical quality-of-life scores improved clearly over 72 weeks. The Mental Component Summary did not significantly improve in either arm, a result the paper's own conclusion omits.

## Key points

- Mental Component Summary did NOT significantly improve in either arm over 72 weeks.
- Physical scores improved decisively, and more in those who lost more.
- The paper's own conclusion says 'HRQoL improved' and never mentions the mental null.
- Tirzepatide's General Health edge is real but ~1/8 of a standard deviation.
- On-treatment analysis: everyone who stopped is absent from both arms.

SURMOUNT-5 is [the head-to-head trial](https://glpwatchdog.org/research/the-head-to-head-and-what-it-settles), and its quality-of-life analysis found what you would expect on the physical side: people felt better in their bodies, more so the more weight they lost. On the mental side it found something else. The Mental Component Summary **did not significantly improve in either treatment arm** over 72 weeks.[1]

## What was measured

Participants completed the SF-36 version 2 (a widely used health survey that produces a Physical Component Summary, a Mental Component Summary, and eight separate domain scores), along with the PHQ-9 depression questionnaire and a global impression of physical activity, at week 72. Scores are norm-based, meaning the general population averages 50 with a standard deviation of 10, which makes the numbers comparable across studies.

One design choice shapes everything below: this was an *on-treatment* analysis, restricted to people who received the drug at their maximum tolerated dose. Anyone who could not stay on either medication is not in it. That flatters both arms, so keep it in mind before reading any of these as what a randomly chosen person would experience.

## The physical side

*SF-36v2 results at week 72. Norm-based scores; population mean 50, standard deviation 10.[1]*

| Measure | Result |
| --- | --- |
| Physical Component Summary | Improved from baseline on both drugs (p < 0.001) |
| **Mental Component Summary** | **No significant improvement on either (p > 0.05)** |
| All eight domain scores | Improved (p ≤ 0.008) |
| General Health, tirzepatide vs semaglutide | 5.45 vs 4.20 (p = 0.003) |

Physical function, bodily pain, role limitations and vitality all improved, and improved more in people who lost more weight. That picture is coherent and unsurprising: carrying less weight makes moving through a day easier, and the survey detected it. Whether a physical-function gain of that kind is [large enough for a patient to notice](https://glpwatchdog.org/research/real-and-below-the-threshold) is a separate question.

The one head-to-head difference was in General Health, where tirzepatide gained 5.45 points against semaglutide’s 4.20. That difference is real (p = 0.003), and it is **1.25 points on a scale whose standard deviation is 10**, roughly an eighth of a standard deviation. It is statistically dependable but clinically modest, and those are two different claims.

> **The advantage was larger where there was more room for it.** Among participants who began with limited physical function, tirzepatide beat semaglutide on the Physical Component Summary, Physical Functioning and General Health (p ≤ 0.025). People with the least physical capacity at baseline had the most to gain, and gained it.

## The result the conclusion does not mention

The paper’s own conclusion opens by saying health-related quality of life improved with both drugs. That is true of the physical measures and it is not true of the mental summary, which the conclusion does not mention.

> After 72 weeks, the Mental Component Summary had not significantly improved on tirzepatide or on semaglutide.

Since we published this, a network meta-analysis of 43 trials and 45,663 participants reached a related conclusion across the whole drug class: no agent improved quality of life beyond the threshold agreed in advance as noticeable to patients (see [what 262 trials say together](https://glpwatchdog.org/research/what-262-trials-say-together)). One trial finding this is easy to dismiss, but forty-three are not. The exception is heart failure, where two trials found [large gains on a symptom scale built for that disease](https://glpwatchdog.org/research/the-one-place-people-felt-better), a different measure.

This matters because of what people are told to expect. Weight loss is routinely sold as a route to feeling better about yourself, and a great deal of marketing in this category trades on exactly that promise. The best-controlled evidence available, a randomized head-to-head with a validated instrument, found large weight loss improving how people’s bodies functioned without a measurable improvement in the mental component.

> **Be careful about what the null does and does not say.** It does not mean nobody felt better mentally; a summary score can stay flat while individuals move in both directions. It does not mean these drugs harm mental health. And the Mental Component Summary of a general health survey is a blunt instrument for mood. What it does say is that a benefit many people expect did not appear where you would most expect to find it.

## What to take from it

- **Expect physical improvement, and expect more of it if you lose more.** That relationship held across both drugs.
- **Do not plan on the drug fixing how you feel about yourself.** If that is the main reason for starting, this is the evidence to weigh.
- **The head-to-head difference is small.** An eighth of a standard deviation on one domain is not a reason to choose between two drugs; tolerability, cost and supply are larger considerations.
- **It is an on-treatment analysis.** The people who stopped are absent from both arms.

For the weight results these scores sit on top of, see [the tirzepatide timeline](https://glpwatchdog.org/research/tirzepatide-results-timeline); for what these drugs do to motivation specifically, [semaglutide and motivation](https://glpwatchdog.org/research/semaglutide-and-motivation).

## Frequently asked questions

### Does losing weight on a GLP-1 improve mental health?

In the SURMOUNT-5 quality-of-life analysis, the Mental Component Summary did not significantly improve in either the tirzepatide or the semaglutide arm over 72 weeks, despite substantial weight loss. Physical scores improved clearly.

### Does that mean the drugs are bad for mental health?

No. It means a benefit did not appear, not that harm did. A summary score can also stay flat while individuals move in both directions, and this instrument is a general health survey rather than a dedicated mood measure.

### Was tirzepatide better than semaglutide for quality of life?

Slightly, on one domain. General Health improved 5.45 points against 4.20 (p = 0.003), a 1.25-point difference on a scale with a standard deviation of 10, so about an eighth of a standard deviation.

### Did people who lost more weight feel better?

Physically, yes. Greater weight reduction tracked with more improvement in physical function, bodily pain, role limitations, general health and vitality, pooled across both drugs.

### Who was included in this analysis?

Only participants who received their drug at the maximum dose they could tolerate (an on-treatment analysis). People who could not stay on either medication are not represented, which flatters both arms.

## References

1. Shukla AP, Dunn JP, Gomez Valderas E, et al. Improved health-related quality of life with tirzepatide versus semaglutide in adults with obesity or overweight from the SURMOUNT-5 trial. Diabetes, Obesity and Metabolism. 2026. PMID: 41187971. https://pubmed.ncbi.nlm.nih.gov/41187971/

---

Cite as: GLP Watchdog, Nora Bissett, "The Mental Health Score That Did Not Move", published 2026-08-20, updated 2026-10-30. https://glpwatchdog.org/research/the-mental-health-score-that-did-not-move

Canonical HTML page: https://glpwatchdog.org/research/the-mental-health-score-that-did-not-move

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
